Subscribe
 
 

'Pristine' Forest Teemed With People

19 September 2003 (All day)
Comments

Well connected. Wide roads (red) linked ancient towns more populous than modern Kuikuro villages (bottom).

It's not easy for humans to survive in the Amazon, where floods and barren soils take a heavy toll on crops. The most remote tracts harbor just a few scattered villages, and scientists had thought it was ever so. The discovery of a highly planned network of ancient villages, however, adds to the evidence that parts of the Amazon were densely populated and that much of the forest has been deeply impacted by humans.

For more than 20 years, researchers have been building the case that pre-Columbian peoples had a significant impact on much of the Amazon. There were many hints: large mounds near the mouth of the river, sophisticated dikes and fish weirs in Bolivia, forests planted with palms and fruit trees, and patches of soils enriched by humans. But many experts assumed that larger, more complex societies were restricted to the relatively fertile floodplains in the region's lowest lying lands.

That's not the picture emerging from the highlands of the Upper Xingu River, where archaeologist Michael Heckenberger of the University of Florida, Gainesville, has spent more than a decade mapping earthworks such as roads up to 50 meters wide. The newest work, described in the 19 September issue of Science, has revealed 19 pre-Columbian villages linked to smaller settlements by remarkably straight roads. Shorter avenues headed out toward the same points of the compass, as if the villages were all built from a similar blueprint. "This really blew us away," Heckenberger says, for it implied a society much larger and more complex than any in the Amazon today.

Indeed, based on ceramic fragments and other information, Heckenberger's team estimates that each prehistoric cluster supported between 2500 and 5000 people--a much higher figure than usually imagined for upland sites. These ancient farmers also left their mark on the land, Heckenberger argues. Instead of virgin forest, the team found large patches of secondary regrowth. "The point is that in 1492, human influence had spread to essentially the entire area," Heckenberger says. "None of the area was natural."

The networked villages are "an incredibly important indicator of a complex society," says Susanna Hecht, a geographer at Stanford University's Center for Advanced Study in the Behavioral Sciences. "The extent of population density and landscape domestication is extraordinary." The research could also provide insights into how to farm sustainably in the Amazon, because the large populations described by early explorers declined from introduced disease, not agricultural failures.

Related site
Heckenberger's site on the Upper Xingu project

Posted In: